SmartCache: An Optimized Content Caching System API

Overview

SmartCache is a robust and efficient content caching system built using FastAPI. It is designed to cache responses from a simulated slow API and serve cached content with features like cache invalidation, stampede prevention, and statistics tracking. The system ensures optimal performance, thread safety, and efficient memory usage through a Least Recently Used (LRU) eviction policy.

Features

• GET Endpoint: Retrieves content using the cache when available.
• Simulated Slow API: Mimics real-world scenarios with a 2-second delay.
• In-Memory Cache: Includes a default Time-to-Live (TTL) of 5 minutes.
• Cache Invalidation: Supports forced cache refresh for specific content.
• Stampede Prevention: Uses locks to handle concurrent requests efficiently.
• Cache Statistics: Tracks hit rate, miss rate, and current cache size.
• Settings Update Endpoint: Allows TTL and cache size adjustments at runtime.
• Basic Eviction Policy: Implements LRU eviction to manage cache size.
• Thread-Safe: Ensures safe concurrent access to the cache.

Installation

Prerequisites

• Python 3.8+

Setup

1. Clone the repository:

   git clone https://github.com/skidrow8852/smartcache.git
   cd smartcache

2. Create a virtual environment and activate it:

   python -m venv venv
   source venv/bin/activate   # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Run the application:

   fastapi dev

Usage

Endpoints

1. Retrieve Cached Content

GET /content

• Retrieves content from the cache or fetches it from the simulated slow API.

2. Invalidate Cache

POST /invalidate

• Body Parameter:

  {
    "key": "content"
  }

• Invalidates a specific cache entry by key.

3. View Cache Statistics

GET /stats

• Returns statistics such as cache hits, misses, current size, and TTL.

4. Update Cache Settings

POST /settings

• Body Parameter:

  {
    "ttl": 300,
    "max_size": 100
  }

• Updates cache TTL and maximum size dynamically.

Simulated Slow API

The application simulates a slow API that introduces a 2-second delay before returning data. This ensures a realistic caching scenario.

Configuration

The cache has default settings:

• TTL: 5 minutes (300 seconds)
• Max Size: 100 entries

You can update these settings dynamically via the /settings endpoint.

How It Works

1. Cache Retrieval: The system first checks the cache for the requested key.
2. Stampede Prevention: If the cache is empty, a lock prevents concurrent requests from overloading the slow API.
3. Cache Update: Responses are stored in the cache with a timestamp.
4. LRU Eviction: When the cache reaches its maximum size, the least recently used entry is removed.